Thorax 1991;46:700-705

Effects of histamine, ethanol, and a detergent onexudation and absorption across guinea pig airwaymucosa in vivo

Lennart Greiff, Ingrid Erjefalt, Per Wollmer, Ulf Pipkorn,* Carl G A Persson

AbstractThis study examined effects of three sub-stances that cause mucosal provocation(histamine, ethanol, and the detergentdioctylsodium sulphosuccinate (DOSS))on the flux of solutes across airway vas-cular mucosal barriers in anaesthetisedguinea pigs. The inward flux was assessedas absorption of iodine-131 labelledalbumin (MW 69 000) from the tracheo-bronchial surface into the circulationand the outward flux as the exudation oftwo intravenously administered plasmatracers-'25I albumin (MW 69 000) andfluorescein isothiocyanate conjugated(FITC) dextran (MW 70 000)-into theairway. The absorption of technetium-99m labelled DTPA (MW 492) from thetracheobronchial airways was deter-mined in separate experiments. Hista-mine (5 0 nmol) dissolved in 40 I1 salineand superfused on to the tracheobron-chial mucosal surface caused significantand similar entry of 125I albumin andFITC dextran into the airway lumen.This dose of histamine did not, however,alter the absorption of small ("mTcDTPA) or large ('31I albumin) solutesacross the airway mucosa. Ethanol (0 17ymol), superfused in the same way, alsocaused significant exudation of theplasma tracers into the airway lumen. Inaddition, ethanol increased the absorp-tion of "'lI albumin without causingchange in the disappearance rate of "mTcDTPA. The detergent, DOSS (0f28nmol), dissolved in ethanol (0-17 pmol),caused a pronounced increase in exuda-tion and much increased absorption ofsmall and large tracer solutes. Thusthree patterns of change in airwaymucosal barriers were found. The agentsthat are toxic to membranes, ethanoland DOSS, caused a bidirectional in-crease in permeability across themucosa, whereas histamine caused onlyan outward exudative flux. The resultsobtained with histamine are similar tothose seen previously with bradykinin,capsaicin, and allergen, suggesting thatendogenous inflammatory mediatorshave a role in mucosal defence, produc-ing entry of plasma exudates into theairway lumen without increasing themucosal absorption of luminal material.

Plasma exudation may occur in asthma andcontribute to the pathogenesis of the disease.'We have previously shown that inflammatoryprovocation of the guinea pig tracheobron-chial mucosa causes prompt exudation ofplasma into the airway tissue and lumen.23The entry of substantial amounts of largesolutes from plasma into the lumen shows thatthe exudation permeability of the epitheliallining has greatly increased. It is widelybelieved that such changes in epithelial per-meability are bidirectional and hence thatincreased exudation is associated with in-creased absorption of luminal material acrossthe epithelium."7 Recent in vivo and in vitroobservations in guinea pig airways, however,suggest that the changes associated withinflammation are predominantly exudative.89Mucosal challenge with allergen, bradykinin,and capsaicin caused a 15-20 fold increase inthe luminal content of plasma tracers with noconcomitant increase in the absorption oftracer molecules from the airway lumen.9 Dataobtained in vitro suggest that the unidirec-tional flux of macromolecular solutes into theairway lumen may be via hydrostatic pressureoperated valve like passages between epithelialcells.'0 ' Morphological and functional dataalso suggest that after exudation of unfilteredplasma into the lumen the epithelial liningpromptly resumes its normal tightness.'3Data from nasal studies in man also support

these findings. Svensson et al " found that thenasal mucosa responded to repeated histamineapplications with a reversible and repeatableexudative response even when the intervalbetween the histamine applications was asshort as 30 minutes. Greiff et al 5 showed thatthe change in mucosal permeability wasunidirectional as the absorption of chromium-51 labelled EDTA (MW 372) across themucosa was not increased with prolonged his-tamine induced exudation. Entry of a plasmaexudate into the airway lumen should be con-sidered as a first line defence mechanism of thenormal mucosa,8 in view of the prompt res-ponse to provocation and the non-injuriousand unidirectional nature of the process.3 911 15

In this study we have determined plasmaexudation and mucosal absorption of solutesin guinea pig airways, using different topicalprovocations. Histamine was examined to seeif it produced an increase in exudation with noincrease in solute absorption, as observed withbradykinin, the neurogenic stimulus cap-

Department ofOtorhinolaryngologyL GreiffU Pipkorn*Department ofPharmacology,Astra Draco, Box 37,S-221 00 Lund, SwedenI ErjefiltDepartment ofClinical PhysiologyP WollmerDepartment ofClinicalPharmacology,University Hospital,S-22185 Lund, SwedenC G A PerssonReprint requests to:Dr PerssonAccepted 28 May 1991*Died June 1990.

700

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from

Effects of histamine, ethanol, and a detergent on exudation and absorption across guinea pig airway mucosa in vivo

saicin, and an IgE driven allergic reaction.9The effects of histamine were compared withthose of a detergent, dioctylsodium sulpho-succinate (DOSS), and its vehicle, ethanol.DOSS is likely to increase the absorptionpermeability of the airway mucosa, but itseffect on plasma exudation is unknown. Asprevious studies in guinea pig tracheo-bronchial airways had examined onlymacromolecular absorption, we studied theabsorption of both a large (iodine-131 labelledalbumin, MW 69 000) and a small (techne-tium-99m labelled DTPA, MW 492) luminalsolute.

MethodsSTUDY DESIGNWe examined the outward (exudation) andinward (absorption) permeability of the vas-cular mucosa airway barriers after mucosalprovocation with histamine (5 0 nmol), ethanol(0-17 imol), and DOSS (0-28 nmol) plusethanol (0 17 imol). Fluorescein isothiocya-nate conjugated dextran (FITC dextran, MW70 000) and 125I albumin (MW 69 000) wereadministered intravenously as plasma tracers."'I albumin (MW 69 000) and 99mTc DTPA(MW 492) were superfused on to the trachealmucosal surface as large and small solute ab-sorption tracers respectively in separate groupsof animals.

PREPARATION OF ANIMALSMale guinea pigs weighing about 400 g, andwith a normal weight gain during 10 days'observation, were used. After starvation forfour hours the animals were anaesthetised witha 3:2 mixture of ketamine (50 mg/ml) andxylazine (20 mg/ml) 1 0 ml/kg intramuscularly.These drugs maintain stable anaesthesia dur-ing experiments and exert little effect on arterialblood pressure.'6To prepare for tracheal superfusion, a thin

plastic tube (outer diameter 0-6 mm) wasintroduced into the proximal tracheal lumenvia the modth and positioned with its tip about1 cm below the larynx.' The tracheal catheterwas connected to a servo pump and the animalswere inclined at a 450 angle (head up) tofacilitate tracheobronchial distribution of thesuperfusate.

PLASMA EXUDATION AND ABSORPTION OF131I ALBUMINIntravenous administration ofplasma tracersTen minutes after tracheal catheterisation theplasma tracers FITC dextran (5%) in saline(0 25-030 ml/100 g) and 125I albumin (about0 5 MBq) in saline (0 025-0)05 ml/100 g) wereadministered via a marginal vein in the ear.

Tracheobronchial administrationTen minutes after the administration ofplasmatracers the surface of the tracheal mucosa wassuperfused with solutions (40 pl) delivered at aconstant rate over two minutes. The absorptiontracer "'I albumin (about 20 kBq) was givenwith histamine (5 0 nmol), ethanol (0- 17 pmol),and DOSS (0-28 nmol) plus ethanol (0 17

umol), or saline control. All solutions containedguinea pig albumin (3-0 mg/ml).

After completion of the superfusion thetracheal catheter was withdrawn. The animalswere kept in the same position for a further fiveminutes and then placed on a heating pad in thehorizontal position for five minutes. Each treat-ment group contained six animals. Twosamples (40 M1) ofthe superfusion solution wereobtained as standards and counts were made onthem at the same time as those on the tissue andplasma samples from each animal.

Termination of the experimentsThe experiments were terminated by intra-venous administration of pentobarbital (60mg/ml) 0-2 ml. A thoracotomy was performedand a mixture of arterial and venous blood wascollected through an incision made into theventricles ofthe beating heart. After addition ofheparin the samples were centrifuged (500 g,six minutes) to separate the plasma.

Tracheal lavageThe trachea was lavaged as previously de-scribed and validated.' Briefly, the larynx,trachea, and lungs were excised en bloc. Thepreparation was arranged in a tilted positionwith the larynx upwards and the lobar bronchiwere tied with ligatures at the hilar level. Aplastic tube was introduced through an incisionin the left principal bronchus and secured by aligature. The mucosal surface of the lowerportion of the trachea was irrigated throughthis tube by four retrograde superfusions overone minute with 0-25 ml saline. The lavage fluidwas readily visible through the thin trachealwall; the fluid was infused to 1 cm below thelevel where the tracheal tube had beenpositioned.

Tracheal tissue samplingThe trachea was cleaned macroscopically fromexternal vessels and lymph nodes. A section ofthe superfused and lavaged intrathoracic partof the trachea was excised for tracer analyses.

Analyses of exuded 125I albumin and FITCdextran and absorbed 13'I albuminThe samples of plasma, lavage fluid, and tra-cheal tissue were weighed. The activities of 125Iand "' I were counted in a well counter(Harshaw). The detector was connected to adual channel analyser (LEAB) and computer.Samples from each animal were counted for 60seconds. The propagation of 1251I activity into13"I was less than 0 1% whereas the propagationof "'I activity into 125I was about 20%. Toavoid the influence of scattered radiation there-fore the activity of I3"I was counted first and theactivity of 1251I 20 days later (the half lives are:"'I eight days, 125I 60 days). FITC dextran insamples of plasma, tracheal lavage fluid, andcontrol solutions were analysed by fluorimetry(Perkin Elmer LS-5).

CalculationsThe exudation of plasma was calculated bydividing the plasma tracer values for lavage and50 mg tracheal tissue samples by the values in

701

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from

Greiff, Erjefalt, Wollmer, Pipkorn, Persson

1-0 g plasma. Exuded plasma was expressed inmicrolitres ofplasma in 50 mg oftracheal tissueand recovered lavage fluid with reference to thecorresponding plasma tracer. The density ofplasma was set at 10. The absorption of '"'Ialbumin was calculated by dividing the radio-activity in 1-0 g of plasma by that obtained insamples of the superfusate, expressed as afraction of the total amount of activity super-fused through the tracheal catheter. Theamount of '"'I albumin was also determined in50 mg tracheal tissue samples.

Stability of the tracersIntravenous 13'I albumin is stable in the guineapig tracheal superfusion model during thecourse of the experiment both in the circulationand after exudation into the airway lumen.3We used the same gel filtration technique(Sephadex G25, Pharmacia, Uppsala,Sweden)3 to examine the binding of 131I toalbumin both in the superfusate (containing "'lIalbumin, DOSS 0-28 nmol, and ethanol 0-17,mol) and, after its absorption, in plasmarecovered 10 minutes after the tracheal super-fusion (n = 4). Samples of the superfusate(containing about 5 x 105 counts/min) and1 ml plasma (containing 700-900 counts/min)diluted in saline were filtrated. The mean 13'Iactivity detected in the albumin fractions of thelavage fluid and the plasma samples was 93%(SE (3%)) and 98% (1%). FITC dextran isstable during prolonged in vitro and in vivoexperiments."3

TRACHEOBRONCHIAL DISTRIBUTION OF THESUPERFUSATE AND ABSORPTION OF 99mTc DTPATracheobronchial administration of99'FTc DTPA and active agentsTen minutes after tracheal catheterisation theanimals were inclined to a 450 angle (head up)and positioned under a gamma camera. Thesurface of the tracheal mucosa was superfusedwith solutions (40 ,ul) delivered at a constantrate over two minutes. The solutions contained99mTc DTPA (about 8 MBq) in saline andhistamine (5 0 nmol), ethanol (0 17 umol), andDOSS (0 28 nmol) plus (ethanol 0 17 4umol) orsaline control. After superfusion the trachealcatheter was withdrawn and the animals werekept in the inclined position for measurementof the disappearance of `smTc DTPA. Eachgroup contained six animals.

Exuded plasma (pi) in tracheal lavage fluid calculatedfrom the detection offluoresceinisothiocyanate conjugated (FITC) dextran and 125I albumin

Mean (SEM) exuded plasma (pl)

Method: Control Histamine Ethanol DOSS plus ethanol

FITC dextran 0 13 (004) 1 31 (038)** 0-76 (0 25)* 9.55 (1-53)***'"'I albumin 0-15 (004) 1-57 (0-41)** 0-80 (0-23)* 8-11 (1-66)***

Histamine, ethanol, and dioctylsodium sulphosuccinate (DOSS) plus ethanol caused significantexudation of plasma. Comparisons are made with saline control data (*p < 0 05, **p < 0-01,***p < 0-001: Student's t test). There was a high correlation between individual measurements

of FITC dextran and 12 I albumin.

Distribution of the superfusate and absorption of99'nTc DTPAThe distribution and absorption (dis-appearance) of SSmTc DTPA were monitoredcontinuously for 20 minutes with a gammacamera (Portacamera, General Electric) equip-ped with a converging collimator. Images wereobtained in successive one minute frames andstored in a 64 x 64 image matrix. A region ofinterest was selected over the tracheobronchialairways. Time-activity curves were generatedand the half life ofdisappearance was calculatedafter a monoexponential equation had beenfitted to the experimental data by the leastsquares method. A region of interest was alsoselected just proximal to the instillate to assesstransport of the administered tracer by muco-ciliary activity.

DRUGSThe anaesthetics were ketamine (Ketalar 50mg/ml, Parke-Davies), xylazine (Rompun vet20 mg/ml, Bayer), and pentobarbital(Mebumal vet 60 mg/ml, Nordvacc). Thetracers were 125I and 131I labelled human serumalbumin (Disab), fluorescein isothiocyanateconjugated dextran (Bioflor), and `smTc DTPA(Amersham International). The chemicalswere guinea pig serum albumin, histamine, anddioctylsodium sulphosuccinate (Sigma).

ANALYSISAnalysis of variance was used to examine thedifferences in exuded plasma tracer levels (1251albumin and FITC dextran), absorbed "3'Ialbumin levels, and half lives of the `SmTcDTPA time-activity curves. If changes werestatistically significant further analyses werecarried out with Student's t test for unpairedobservations. Values of p less than 0 05 wereconsidered significant (two tailed test). Aregression analysis of individual values wasused to examine the correlation betweenmeasurements of lavage fluid plasma as deter-mined by 1251I albumin and by FITC dextran.Values are presented as means with standarderrors in parentheses.

ResultsPLASMA EXUDATIONHistamine, ethanol, and DOSS plus ethanolproduced significant exudation of plasma intothe tracheal tissue and lumen (table, fig 1).There was a high correlation between 125Ialbumin and FITC dextran in the lavagesamples (r = 0-94).

131I ALBUMIN ABSORPTIONHistamine did not increase the absorption of131I albumin into the circulation. In contrast,ethanol increased the absorption of 13 I albuminand this effect was much greater with DOSSdissolved in ethanol (fig 2). The content of 131Ialbumin in the tracheal tissue samples was onlymarginally affected by these treatments. Thetissue levels (expressed as the fraction x 10-3 ofthe total amount of instilled activity detected in50 mg tracheal tissue) were: control 6 4 (1 9),

702

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from

703Effects of histamine, ethanol, and a detergent on exudation and absorption across guinea pig airway mucosa in vivo

r

-I-

Control Histamine Ethanol

to the oesophagus cannot, however, be ex-cluded. The count rate over the instillate

** *w decreased with time but there was no visiblechange in the shape of the boli in either group.In the group receiving DOSS plus ethanoldiffuse tissue background activity appeared,and radioactivity could be seen to accumulatein the bladder. No proximal transport of thesuperfusate in the trachea was detected. Theretention curves obtained after treatment withsaline, histamine, and ethanol were similar(p > 0 05, analysis of variance), whereas theabsorption of99mTc DTPA was much increasedin the group given DOSS and ethanol (p <0 001, comparison with results for saline) (fig4). The half lives of the tracer in the airwayswere: control 58 6 (4 5), histamine 55 4 (3 3),ethanol 55-7 (4 1), DOSS plus ethanol 64 (10)minutes.

DOSS

Figure 1 125I albumin recovered in tracheal lavage fluid (solid columns) and 50 mgtracheal tissue (hatched columns) after control, histamine, ethanol and dioctylsodiumsulphosuccinate (DOSS). *p < 0-05, **p < 0 01, ***p < 0 001 (comparisons withcontrol values).

histamine 10-0 (2 4) (p > 0-05); ethanol: 6-8(2-8) (p > 0 05); and DOSS plus ethanol: 5 7(1 5) (p > 0-05). The comparisons are madewith the controls.

TRACHEOBRONCHIAL DISTRIBUTION AND 99mTcDTPA ABSORPTIONMonitoring of 99mTc DTPA indicated a localand, during the course of the experiment, welldefined distribution of the fluid in the tracheaand proximal bronchi. Neither histamine nor

ethanol altered the distribution of the boli (fig3); DOSS plus ethanol largely maintained thesame distribution of the boli but there was

some diffusion upwards in the trachea im-mediately after its administration (fig 3). With a

test area selected over the gastrointestinal tractwe could not detect any swallowed activity inany experiment. Some displacement of activity

0

Exv) >M +j

C.)

_E ._0

>i (m 4-

.0i- 400C.Of.n L.Q4 4

3-

2-

-

0 -

*T

Control Histamine Ethanol DOSSFigure 2 Absorption of '"'I albumin into circulating plasma (10 g), expressed as a

fraction of the total activity superfused on to the mucosal surface, after control,histamine, ethanol, and dioctylsodium sulphosuccinate (DOSS). **p < 0-01,***p < 0-001 (comparisons with control values).

DiscussionBy using 99mTc DTPA and external gammacamera detection we have confirmed previousobservations39 12 17 that the primary distribu-tion ofthe tracheal superfusate is confined to thecentral airways and that the agents in thesuperfusate that caused exudation did notchange this distribution appreciably during theexperiment. Our present superfusion tech-nique thus permits the administration of con-

trolled doses of agents to a rather well definedarea of the tracheobronchial mucosal surface. Afurther aspect of the present method is thatboth the provoking agent and the absorptiontracer were given in the same superfusate, thusensuring that they have a similar airway dis-tribution.Only small amounts of the exudation tracers

FITC dextran and 125I albumin were found inthe airway lumen of control animals. Thisresult agrees with previous findings, showingalso that none of the larger plasma tracer, 131Ifibrinogen, entered the airway lumen. Thesedata confirm the non-traumatic nature of thetechniques we used for the mucosal challengeand subsequent retrieval of mucosal surfaceliquids.3 We have previously shown that aftercirculation in plasma and exudation into theairway lumen over 25 minutes the binding of131I to albumin is at least as tight as it was in thestandard solution injected intravenously.3 Thepresent gel filtration results show equally goodstability for 131I albumin after it had beensuperfused on the airway mucosal surface andabsorbed into the circulation.Histamine increases exudation of plasma

into the airway lumen in a dose dependentmanner.' This exudation with histamine (5 0nmol) has now been confirmed with bothcharged (125I albumin) and uncharged (FITCdextran) plasma tracers. The close correlationbetween the two tracers in the exudate (table)confirms previous data,23 and support the viewthat the entry ofalbumin into the airway lumenin response to inflammation is largely an unfil-tered flow ofplasma across the airway endothe-lial-epithelial barriers rather than a secretoryevent.2'The 10 fold increase in 125I albumin exuda-

-oC.5

-

0 ns

> Q)

.-ac. m

-XX0 u

(u)

20 -

10 -

0-

.I

0

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from

704Greiff, Erjefalt, Wollmer, Pipkorn, Persson

Figure 3 Gamma cameravisualisation of the "'TcDTPA superfusate inguinea pigtracheobronchial airways.The dark areas show thedistribution of 99*TcDTPA in thetracheobronchial airwaysimmediately after thesuperfusion (left panels)and after 20 minutes(right panels) in groupsreceiving saline (A),histamine (B), ethanol(C), and dioctylsodiumsulphosuccinate (DOSS)plus ethanol (D). Thesurrounding line shows thecontour of the animal. Adiffuse tissue backgroundgradually appeared in allgroups, especially thegroup receiving DOSSplus ethanol; but there wasessentially no change inshape of the airway boli.The animals receivingDOSS plus ethanol (D)showed pronouncedradioactivity in thebladder after 20 minutes.

tion in response to histamine was not associatedwith any increased absorption of the luminal"'lI albumin tracer. In this respect histaminemimicks the action of bradykinin, capsaicin,and allergen in guinea pig tracheobronchialairways.9 This is the first study to use histamineand to employ exudation and absorptiontracers in the same animal.Our study shows that during exudation the

absorption barrier is also maintained for smallsolutes such as "mTc DTPA (MW 492). Thisalso occurs in human nasal airways, where anexudative dose ofhistamine had no effect on theabsorption of 5"Cr EDTA (MW 372).'5 It maybe a general property of endogenous exudativemediators to enable exuded plasma to enter the

.i 585

5651 _

2

airway lumen without compromising the air-way mucosa as a barrier to luminal solutes. Theunidirectional change in epithelial perme-ability would allow plasma exudate to operateas a first line defence mechanism on the surfaceof a mucosa that remains a functional barrier toabsorption of luminal material.Boucher et al5 found that an aerosol of

histamine increased the rate of absorption ofhorseradish peroxidase (MW 40 000) inguinea pigs. The tracer was given as a trachealinstillate in a volume about five times largerthan the present superfusate. More recently,Ranga et al7 found less consistent effects ofallergen on the absorption of horseradishperoxidase from guinea pig lower airways, but

3334

u sA C

w 511 2

B D

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from

Effects of histamine, ethanol, and a detergent on exudation and absorption across guinea pig airway mucosa in vivo

100-

80 -

-W 60-

0Z. 40-0

20 -

n Ilvp

0 1 0Time (min)

Figure 4 Time-activity curves from the tracheobronchial airways of animals receivingsaline, histamine, ethanol (upper three lines), and dioctylsodium sulphosuccinate(DOSS) plus ethanol (lower line).

concluded that absorption was increased. Asthe airway distribution of histamine, allergen,and horseradish peroxidase in these studies isunknown, making direct comparisons with thepresent findings is difficult. When we used oursuperfusion technique to administer a mus-carinic agonist (carbachol) together with anabsorption tracer we could not confirm thereport that muscarinic agonists increase theabsorption permeability of the airways.5 Thusour present and previous data9 suggest thatneither mediators that cause exudation (his-tamine, bradykinin, capsaicin, and allergen)nor those that do not (muscarinic agents)increase tracheobronchial absorption ofluminal solutes.Histamine is an endogenous agent that may

be released in airway defence operations,8whereas DOSS and ethanol may be consideredas agents with non-specific toxic effects on theairway mucosa. DOSS may be similar in itsaction to large doses of another detergent,lysophosphatidylcholine, which is known tocause severe damage to the mucosa." In ourstudy ethanol, used as a vehicle for DOSS,caused exudation of plasma. It also increasedthe absorption of "3'I albumin but did not affectthe rate of disappearance of 99"Tc DTPA,indicating that it may increase macromolecularabsorption across the mucosa specifically inaddition to causing exudation of plasma. Themechanism underlying the present size depen-dent permeability changes remains unknown,but a distinction between small and large solutepermeability has previously been observed inintestinal mucosa. 9 DOSS (given with ethanol)caused pronounced increases in both theexudation (outward) and absorption (inward)permeabilities of the airway mucosa. Threepatterns of changes in the mucosal barrierfunctions were thus recorded: histamineproducing exudation without increasing ab-sorption, ethanol producing exudation andselectively increasing large solute absorption,and DOSS plus ethanol producing massiveexudation and large increases in small and largesolute absorption. The tissue response com-mon to the three provoking stimuli was exuda-

tion of plasma, which is consistent with theview that this response is a primary defencemechanism.8 Further studies are warranted to

= determine the mechanisms concerned in the'- abnormalities of increased airway absorption

observed with ethanol and DOSS.In conclusion, this study has shown that

histamine produces exudation of plasma intothe airway lumen with no concomitant increasein the absorption of small and large solutesfrom the lumen. This appears to be a specific

E-I action of endogenous mediators because toxicinflammatory factors produced changes in theabsorption permeability of the mucosa in addi-tion to increased exudation. These observa-tions support the view that a main purpose of

_, autacoid induced plasma exudation responses2 0 in the airways is to allow the potent protein

systems of plasma to carry out defence opera-tions on the surface of an intact airway mucosa.This study was supported in part by grants from the SwedishMedical Research Council (projects 8308 and 2872), theSwedish Association Against Heart and Chest Diseases, and theTorsten and Ragnar Soderberg Foundation.

1 Persson CGA. Role of plasma exudation in asthmaticairways. Lancet 1986;ii:1126-9.

2 Persson CGA, Erjefilt I. Inflammatory leakage of macro-molecules from the vascular compartment into thetracheal lumen. Acta Physiol Scand 1986;126:615-6.

3 Erjefilt I, Persson CGA. Inflammatory passage of plasmamacromolecules into airway wall and lumen. PulmPharmacol 1989;2:93-102.

4 Leskowitz S, Salvaggio JE, Schwartz HJ. An hypothesis forthe development of atopic allergy in man. Clin Allergy1972;2:237-42.

5 Boucher RC, Ranga V, Pare PD, Inoue S, Moroz LA, HoggJC. Effect of histamine and methacholine on guinea pigtracheal permeability to HRP. J Appl Physiol 1978;45:939-48.

6 Hogg JC. Bronchial mucosal permeability and its relation-ship to airway hyperreactivity. J Allergy Clin Immunol1981;67:421-7.

7 Ranga V, Powers MA, Padilla M, Strope GL, Fowler L,Kleinerman J. Effect of allergic bronchoconstriction onairways epithelial permeability to large polar solutes in theguinea pig. Am Rev Respir Dis 1983;128:1065-70.

8 Persson CGA, Erjefilt I, Alkner U, Baumgarten C, GreiffL,Gustafsson B, et al. Plasma exudation as a first linerespiratory mucosal defence. Clin Exp Allergy 1990;21:17-24.

9 Erjefilt I, Persson CGA. Allergen, bradykinin, and cap-saicin increase outward but not inward macromolecularpermeability of the guinea pig tracheobronchial mucosa.Clin Exp Allergy 1991;21:217-24.

10 Persson CGA, Erjefilt I, Gustafsson B. Subepithelialhydrostatic pressure may regulate plasma exudationacross the mucosa. Int Arch Allergy Appl Immunol 1990;92:148-53.

11 Gustafsson B, Persson CGA. Asymmetrical effects ofincreases in hydrostatic pressure on macromolecularmovement across the airway mucosa. Clin Exp Allergy1991;21:121-6.

12 Persson CGA, Erjefilt I, Andersson P. Leakage of macro-molecules from guinea pig tracheobronchial micro-circulation. Effects of allergen, leukotrienes, tachykinins,and anti-asthma drugs. Acta Physiol Scand 1986;127:45-105.

13 Luts A, Sundler F, Erjefilt I, Persson CGA. The airwayepithelial lining in guinea pigs is intact promptly after themucosal crossing of large amount of plasma exudate. IntArch Allergy Appl Immunol 1990;91:385-8.

14 Svensson C, Baumgarten CR, Pipkorn U, Alkner U, PerssonCGA. Reversibility and reproducibility of histamineinduced plasma leakage in nasal airways. Thorax 1989;44:13-8.

15 Greiff L, Wollmer P, Pipkom U, Persson CGA. Absorptionof" Cr EDTA across the human nasal airway barriers inthe presence oftopical histamine. Thorax 1991;46:000-00.

16 Hart MV, Rowles JR, Hohimer AR, Morton MJ, HosenpudJD. Hemodynamics in the guinea pig after anesthetizationwith ketamine/xylazine. American Journal of VeterinaryResearch 1984;45:2328-30.

17 Greiff L, Wollmer P, Erjefalt I, Pipkom U, Persson CGA.Clearance of"'Tc-DTPA from guinea pig nasal tracheo-bronchial and bronchoalveolar airways. Thorax 1990;45:841-5.

18 Lindahl M, Hede AR, Tagesson C. Lysophosphatidyl-choline increases airway and capillary permeability in theisolated perfused rat lung. Exp Lung Res 1986;1:1-12.

19 Westrom BR, Tagesson C. Low molecular weight tracersdo not reflect intestinal macromolecular permeability.J Pedriatr Gastroenterol Nutr 1989;8:422.

705

on April 17, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.10.700 on 1 O

ctober 1991. Dow

nloaded from